Internal states of pulse-type hardware neuron model with chaotic dynamics
Kazutaka Someya, Hidekazu Shinozaki, E. Miyata, Y. Sekine · 2002
Information processing in living organisms is carried out by the nervous system. From an engineering viewpoint, the nervous system has various interesting features. So, it is necessary to study the hardware neuron model for its engineering applications. On the other hand, it is suggested that a new information processing structure can be realized in a neural network with a neuron model with chaotic dynamics. We discuss the internal states of a pulse-type hardware neuron model ("P-HNM"), and its response characteristics to periodic pulses. As a result, it is verified that the P-HNM has the properties of the chaotic neuron model, it also exhibits chaos phenomena, resembling the property of actual neurons, and it is available as the chaotic neuron model.